IL-15, that activates natural killer (NK) cells, has potential applications in cancer immunotherapy. Based on rhesus macaques' experiments, we performed a first-in-human phase I dose-escalation trial of Escherichia coli (E. coli) produced rhIL-15 with NCI as the sponsor, Dr. Kevin Conlon as the Principal Investigator (PI) and Dr. Thomas Waldmann as Study Chairman and IND holder. This study was completed with 5 patients having finished their course of therapy at 3.0 mcg/kg/day, 4 patients having entered at 1.0 mcg/kg/day and 5 patients at the 0.3 mcg/kg/day dose of 12 daily infusions. These bolus infusions were associated with marked alterations in numbers and state of activation of multiple lymphocyte subsets that ultimately yielded a significant expansion of NK cells with a less significant expansion of CD8+ effector T cells. Lymphocytopenia preceded target-cell expansion with NK and memory T-cells virtually completely disappearing from the circulation within minutes upon IL-15 administration suggesting localization to tissues. Among CD8 T-cells, effector memory cells responded the fastest followed by central memory, stem cell memory and naive T cells. By the second day of IL-15 administration NK cells began to reappear in the circulation and expanded above baseline. Subsequently IL-15 treatment produced an 8.2-fold expansion of NK cells and a twofold expansion of CD8+ CD45RO+ memory T-cells. This expansion was preceded by increased proliferation quantitated by Ki67 expansion and by activation as defined by expression of CD38 and HLA-DR markers. Expansion was evident by day 8 with maximum increases seen at day 14 or 21 timepoints for NK and CD8+ CD45RO+ cells. NK cells increased in numbers to eightfold over baseline. The primary objectives of this phase I first-in-human study of bolus intravenous recombinant IL-15 (rhIL-15) were to determine toxicity profile, safety, dose-limiting toxicity and maximum tolerated dose of intravenous rhIL-15 administered as daily intravenous bolus infusions for 12 consecutive days in subjects with metastatic malignant melanoma and metastatic renal cell cancer. The first patient treated developed hypotension (B/P: 78/42 mm/Hg) approximately 5 hours after the first dose. The fifth patient at 3.0 mcg/kg/day dose developed dose-limiting thrombocytopenia after six doses. After these patients developed a DLT the Protocol was amended to add two lower dose levels of 1.0 and 0.3 mcg/kg/day. Two of four patients manifested dose-limiting toxicities (abnormalities of liver function) at the 1.0 mcg/kg/day dose. All patients with IL-15 administered at 0.3 mcg/kg/day dose received all 12 doses without DLT. Therefore the maximum tolerated dose of bolus IL-15 was deemed to be 0.3 mcg/kg/day for 12 days of bolus infusions. There was a temporal pattern of post-treatment adverse events with fever and rigors beginning at 2 and a half to 4 hours following start of rhIL-15 infusions, reliably peaking at 3 to 4-hour timepoints. Patients had elevated blood pressure during the rigor period but hypotension with a nadir of approximately 20 mm/Hg below treatment levels ensued. The initiation of fever was contemporaneous with development of a 30-fold elevation in IL-6 serum concentrations with the 3.0 mcg/kg/day dose. Evidence of anti-tumor activity as assessed by CT scans and determination of clinical response by RECIST criteria showed no responses, with stable disease as the best response. However 5 patients manifested a decrease between 10 and 30 percent in marker lesions; 2 of these 5 patients had clearance of lung lesions. The hypothesis that observed acute toxicities were related to cytokine release is supported by the marked elevations of IFN gamma, IL-6 and IL-8 seen in association with the high initial Cmax IL-15 levels. It is probable that bolus infusion is not the optimal way to provide IL-15. By administering IL-15 by continuous intravenous infusions subcutaneously to rhesus macaques the exceedingly high Cmax observed with bolus infusions was avoided. Furthermore, in contrast to eightfold expansion of CD8 effector memory cells observed with bolus administration, when IL-15 was administered at a dose of 20 mcg/kg/day by continuous intravenous infusion for 10 days to non-human primates there was a massive (80-100-fold) expansion of circulating CD8 effector memory T-cells. To translate these observations into clinical trials with the goal of reducing Cmax, the excess cytokine release and of maintaining optimal IL-15 levels for a long period, we have initiated a phase I trial of rhIL-15 administered by continuous intravenous infusions to patients with metastatic malignancy. Furthermore, we joined with the Cancer Immunotherapy Trials Network (CITN) in a phase I trial to determine the safety, toxicity profile, DLT and MTD of s.c. rhIL-15 given for 5 days a week for 2 weeks. Although IL-15 administration may show efficacy in the treatment of the metastatic malignancy in human trials, it is unlikely to be optimal when used as a single agent. IL-15 is normally associated with IL-15R alpha. Indeed, IL-15 and IL-15R alpha may be considered a single heterodimeric cytokine. We are using two strategies to circumvent the very low expression of IL-15R alpha seen in normal individuals and patients with malignancy. In the first strategy we are developing IL-15 pre-associated with IL-15R alpha or with IL-15R alpha IgG1-Fc agents that showed enhanced activity on proliferation of CD8+ CD44hi T cells and that demonstrated augmented antitumor efficacy in murine tumor models. We also developed an alternative strategy to augment DCs expression of IL-15R alpha. In this approach an agonistic anti-CD40 antibody was administered to mice in conjunction with IL-15 to circumvent the problems discussed with IL-15 when used in monotherapy. We demonstrated that this combination increased expression of IL-15R alpha on splenic DCs and provided dramatic synergistic therapeutic effects when compared to either agent administered alone. In our future plans we wish to translate preclinical studies by altering the dosing strategy of rhIL-15 by replacing bolus infusions with subcutaneous or continuous intravenous infusions thereby reducing the Cmax and prolonging the period when IL-15 is at the optimal concentration. Furthermore we plan to replace IL-15 monotherapy with IL-15R alpha/IL-15 or IL-15R alpha IgG-Fc/IL-15 or alternatively with IL-15 given in conjunction with an agonistic anti-CD40 monoclonal antibody that augments DCs IL-15R alpha expression and thereby IL-15 action. It is hoped that with these approaches we can reduce the toxicity, increase the impact of IL-15 on activation and expansion of lymphoid populations and thereby improve antitumor effects of rhIL-15 in patients with refractory and metastatic malignancies.